The Cost of Bad Code
There are different ways to measure bad code but, as Associate Justice Stewart of the U.S. Supreme Court back said in 1964, ‘I know it when I see it.’
Developers who have been active for even a mere two years are familiar with messy code. They see how hard it can be to deliver a new feature or a bug fix when the existing code is difficult to disentangle. In those circumstances, a change to one part of the code can break several other places, and sometimes there is no trivial change that can be made without a moderate or high degree of risk.
Over time, messy code gets messier. The problem with deteriorating code is that future enhancements and bug fixes take longer and longer to deliver. The development team’s velocity slows to a crawl until eventually it hits a breaking point.
It is clear that bad code has a high cost. There is a spectrum of solutions available from doing nothing about it, to a clean whiteboard redesign. Fortunately, for many clients, there are point solutions available in between that will address the mess piecemeal, which will deliver more robust solutions, improve the agility of the organization to respond to change, and reduce the operational risk that arises from legacy software.